Interestingness Research of Association Rules in Incremental Mining Data
Bin Yang · Computer Technology and Development · 2009
The general association rules discovery algorithm is using in the framework of confidence and support.However,in incremental data mining process,such algorithms are required support and confidence to constantly changing.Makes inefficient algorithm itself and the lack of persuasive,such as Apriori algorithm.In this paper,in order to solve such problem,use interestingness framework in incremental degrees of data mining association rules,based on the comparative degree of support,confidence in the framework of the algorithm(such as Apriori,FUP algorithm) and the algorithm based on the degree of interest between the gifted disadvantage.The experimental results show that: Interestingness be able to effectively filter association rules,the association rules in incremental data mining to be always less than or equal degree of support,confidence level(Apriori) algorithm excavated rules.